Audit Methodology

How the audit actually runs

Anyone can say your site has problems. We think you deserve to know how we check, what counts as evidence, and how findings get ranked before you pay for any of it. This page is the working standard behind every audit we ship.

Four working principles

Every finding is a test, not an opinion

Each item in an audit report names the check we ran, the output we observed, and the evidence. If we say a page is invisible to AI crawlers, you get the raw HTML and the rendered DOM side by side. If we say a canonical is wrong, you get the curl output. Anything we cannot demonstrate does not ship as a finding.

Evidence or it didn't happen

Screenshots, response headers, crawl logs, and query-level data travel with every flag. The report is written so your own developer can reproduce any finding in minutes, whether or not you ever hire us to fix it.

Prioritized by money, not by severity theater

A missing canonical on your highest-margin service page outranks a perfect-score failure on a page nobody visits. Every finding carries an impact estimate built from real search volume and CPC data for your market, so the roadmap reads like a business plan, not a linter dump.

No catastrophe script

If the foundation is solid, we say so and hand you the short list that matters. An audit that always finds disaster is a sales tool. Ours is a diagnostic, and sometimes the diagnosis is that you need less work than you thought.

The five phases

01

Access and baseline

Search Console, analytics, and tag access. We snapshot where you stand today: indexed pages, earning pages, query mix, and how AI answer engines currently describe you. Every later claim is measured against this baseline.

02

Technical crawl and render testing

Full crawl of the site, plus the render gap check most audits skip: what your server sends versus what Google, ChatGPT, Claude, and Perplexity actually receive after rendering. Canonicals, hreflang, redirect chains, sitemap and robots hygiene, and Core Web Vitals against current thresholds.

03

Market and SERP reality

We pull live SERP, volume, and CPC data for the queries that matter in your market, then map every important page to one primary term and one intent class. Cannibalization, orphaned money pages, and missing buyer questions surface here.

04

AI search readiness

We test the prompts your buyers actually ask across ChatGPT, Claude, Perplexity, and Google AI Overviews, record who gets cited and why, and review your AI crawler policy: which bots you invite, which you block, and what that choice costs you.

05

Roadmap and walkthrough

Findings ranked by revenue impact and effort into a 90-day plan, delivered with the full evidence pack. You walk through it with the strategist who ran the audit, not a salesperson reading a summary.

What we measure

Standards we hold every site to

These are the same checks we run in our own prospecting. When we tell a business their site is slow or unmaintained, it is because it failed one of these tests, and we can show the output.

We also hold ourselves to them: tanalto.com passes every check on this list, and we published the results of running them on 209 local business websites in our 2026 Local Website Health Study.

See the full audit coverage
  • Server response measured from the field, not a lab guess: slow means slow for your customers
  • Meta and canonical hygiene checked page by page, because templates lie
  • Maintenance signals recorded: a site whose footer still says 2023 tells buyers something, fair or not
  • Performance checked against what real devices tolerate, including image payload and lazy-loading discipline
  • Schema and entity coverage read the way search and AI answer engines assemble it, not the way a validator scores it
  • Measurement verified before it is trusted: tag firing, conversion definitions, and consent-mode gaps

Methodology questions

Is this the same process you use in your free outreach audits?

The quick screens we run before contacting a business use a small public slice of this checklist: server response, title and meta hygiene, canonical presence, maintenance signals, and image payload. The paid audit goes far deeper: full crawl, render-gap testing, SERP and AI-citation research, and a prioritized roadmap.

Who performs the audit?

The same senior strategists who do client work. Nothing is outsourced, and nothing is a tool export with a logo pasted on it.

What do we get at the end?

The 90-day roadmap, your Entity Map, the AI citation inventory, the full evidence pack, and a walkthrough call. Ten business days from access. $1,500 flat, credited to month one if we continue within 30 days.

$1,500 flat. Evidence included.

Roadmap in 10 business days. Fee credited to month one if we continue within 30 days.

Book the Audit